Repository navigation
Prove no subset of ≥97 E8 lines has a rank-8 complex realisation (minimum hitting set 24, spectral + local certificate; no κ) - #6854
Merged
Mister-Meeseeks merged 1 commit intoNov 10, 2026
Conversation
… kappa) Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
No κ claim. This adds
research/p104-e8-rank8-closure/, a registered research package for the cardp104-side-correction. It extends the 14-portHall obstruction of #6846 from the 104 ports of one word to every subset of the 120 E8 lines.
Result. Assume the complex label pattern of #6846: each line
Tcarriesk × dlabel matricesU_T, W_Tover a field, the pairingW_S U_T^Tvanishes on every non-orthogonal pair of lines, and every diagonal block is invertible. Then no set of 97 or more of the 120 E8 lines has a
realisation of inner dimension exactly
8k, over any field. The minimum number of lines meeting every member of the 13-line witness family is exactly 24(the family has 1,209,600 members: an 8-clique of the forced-zero graph plus five pair lines on four of its coordinates, each Hall-deficient with
lam <= 4/5).Proof. (1) The forced-zero graph is SRG(120,56,28,24) and
M = 2A + 8I − JsatisfiesM² = 24M, so any 23-line set contains at least 87 forced-zeroedges. (2) A local lemma at one line (a CNF with 998 variables whose unsatisfiability is certified by a DRAT proof that
drat-trimverifies) says that a hitting set of atmost 23 lines that avoids a line
xcontains at least 12 ofx's 56 neighbours; nine E8 reflections carry it to every line, and a direct run at each of the 120 lines agrees.(3) Counting edges between a 23-line hitting set and its 97 complement lines gives
97·12 = 1164 > 56·23 − 2·87 = 1114. A 24-line set meets every witness with equality in both bounds, so 24 is exact.Checks.
check.py(standard library, about a minute) regenerates the lines, the graph, the spectral identity, all 1,209,600 witnesses with a Hall-deficiency test oneach instance (sha256 pinned), a
k = 1exhaustive control over F2 and F3, the reflections, the local CNF (sha256 pinned) and the retained proof (sha256 pinned), the counting inequality and the24-line set.
./verify.sh --dratalso runsdrat-trimon the retained proof (s VERIFIED).Scope. The label pattern is a hypothesis. Only subsets of the 120 E8 lines with this mask are covered; ranks strictly between
8kand9k, other masks,and sets of at most 96 lines are not covered (a 96-line set avoiding the family exists). Nothing is priced. Open questions are listed in the package.
Prepared by DreamingOfClouds with Claude Opus 5.5 (Anthropic), via Claude Code.
🤖 Generated with Claude Code